Flightbox scorecard · suite: public-demo
4 runs · 2 tasks × 2 agents
Task
browser-use-gpt
claude-cu
checkout-locked-account
auth-wall
pass 0/1 · $0.00
auth-wall
pass 0/1 · $0.08
complete-checkout
pass 1/1 · $0.00
pass 1/1 · $0.36